IoT System Based on Iris Recognition

The Internet of Things is increasingly accessible on the World Wide Web, and WoT provides a new space that makes managing things very easy. The fields of the Internet of Things and because of its many possibilities have been applied very widely in the manufacture and areas of our daily lives. In this paper, it was proposed to use irises to increase data security. The client implements a preprocessing process that consists of the following steps: the first step is to convert color images to the gray level luminosity algorithm, the second step blurring the images by using the median filter algorithm, and step three the images convert to converted to binary images according to a threshold, step four detects the edges by using contour algorithm, and finally resizing the images by using the bilinear interpolation algorithm. The Server cloud side consists of the following steps: the first step is the features extraction which will be used in the sift algorithm to increase the accuracy. The second step uses the eliminated structure of the image and the arrangement of features in the image. The third step uses the k-mean algorithm to extract the algorithm, the fourth step is the classification that uses the algorithm j48, and the algorithm naive Bayes. The Iris recognition dataset is used to evaluate this system, which is the MMU dataset consisting of 450 images of 46 people. The recognition rate for a dataset is 99.99%.